Abstract
Purpose :
In age-related macular degeneration (AMD), retinal pigment epithelial (RPE) cells overexpress pro-angiogenic proteins, such as vascular endothelial growth factor (VEGF), causing the growth of new blood vessels. This neovascularization can be simulated by developing a computational behavioral model to predict the extent and probable location of blood vessel growth based on the production of VEGF by the RPE. Computational modeling allows for more high throughput, time efficient, and inexpensive experiments than in vitro or animal models.
Methods :
The computational model mimics the space between a choroidal parent vessel and the RPE layer as a 2-dimensional matrix. The lower boundary of the matrix represents the RPE layer and serves as a point source of VEGF. The VEGF diffuses towards the upper boundary of the matrix which represents a parent vessel in the choroid from which new blood vessels will grow. Tip cells in the parent vessel react to increasing levels of VEGF by secreting protease. The protease degrades the fibronectin of the extracellular matrix allowing the tip cell to escape the parent vessel and form a new blood vessel. In the computational model, tip cells move in a biased random walk with the probability of moving to an adjacent location based on chemoattractance which is calculated using the concentration of free VEGF, the amount of protease secreted by the tip cell, and the extent of fibronectin degradation. This methodology provides an effective prediction of the extent and location of neovascularization given VEGF production values. These results will then be validated with live imaging HUVEC cell migration studies.
Results :
Our model produces a graphical representation of blood vessel growth, as well as 3D graphs of VEGF, protease, and fibronectin concentrations. Consistent with data found in scientific literature, there is a larger extent of blood vessel growth with higher VEGF expression and new blood vessel growth is biased towards higher concentrations of VEGF. These results will be validated by our HUVEC cell migration studies.
Conclusions :
With the parameter values found through experimentation, the model accurately predicts blood vessel growth over time when RPE cells are overexpressing VEGF in wet AMD. The computational model reduces test variables and provides parameters for lab experiments saving researcher time and money.
This is a 2021 ARVO Annual Meeting abstract.